| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Biochemistry University of Illinois Urbana, Illinois 61801
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
(fER) was translated from a
5'-translation initiation site and fused to the neomycin resistance
gene, which was translated from an internal ribosome entry site. One
stable HeLa-ER-positive cell line (HeLa-ER1) produces 1,300,000
molecules of fER/cell (
20-fold more ER than MCF-7 cells). The HeLa
fER is biologically active in vivo, as judged by rapid
death of the cells in the presence of either 17ß-estradiol or
trans-hydroxytamoxifen and the ability of the cell line to
activate a transfected estrogen response element (ERE)-containing
reporter gene. The FLAG-tagged ER was purified to near homogeneity in a
single step by immunoaffinity chromatography with anti-FLAG monoclonal
antibody. Purified fER exhibited a distribution constant
(KD) for 17ß-estradiol of 0.45
nM. Purified HeLa fER and HeLa fER in crude
nuclear extracts exhibit similar KD values for
the ERE (0.8 nM and 1
nM, respectively), which are approximately 10
times lower than the KD of 10
nM we determined for purified ER expressed
using the baculovirus system. HMG-1 strongly stimulated binding of both
crude and purified HeLa fER to the ERE (KD of
0.25 nM). In transfected HeLa cells, HMG-1
exhibited a dose-dependent stimulation of 17ß-estradiol-dependent
transactivation. At high levels of transfected HMG-1 expression
plasmid, transactivation by ER became partially ligand-independent, and
transactivation by trans-hydroxytamoxifen was increased by
more than 25-fold. These data describe a system in which ER, stably
expressed in HeLa cells and easily purified, exhibits extremely high
affinity for the ERE, and suggest that intracellular levels of HMG-1
may be limiting for ER action. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The toxicity of ER in mammalian cells has made expression of substantial levels of wild-type ER quite difficult, hindering both structural and mechanistic studies of ER action. Although the ER has been expressed in Escherichia coli, yeast, and in the baculovirus system (7, 8, 9, 10), these systems suffer from several problems. While ER fusion proteins have been expressed in E. coli, the level of expression was relatively low (7), and appropriate posttranslational modification of the ER was unlikely to have occurred. ER production in yeast has been hampered by proteolysis and by low levels of functional expression (9). ER expressed and purified in insect cells using the baculovirus system (10, 11, 12) binds to DNA containing the estrogen response element (ERE) poorly (see below), or not at all, unless additional proteins such as HMG-1 are present (11, 12). When ER is expressed in mammalian cell lines, the cells stop growing and lyse after exposure to low concentrations of estrogen (13, 14). Consequently, ER-expressing cell lines have expressed relatively low levels of wild-type ER (15). One approach to circumvent the toxicity of wild-type ER has been to utilize the less toxic ERval400 mutant (16, 17). Using this ER mutant, stable mammalian cell lines have been developed (13, 14, 18, 19, 20, 21, 22, 23, 24, 25, 26), some of which express very high levels of this less toxic mutant ER. While these cell lines will continue to be extremely useful, the ER expressed in these cell lines lacks an epitope tag to facilitate isolation of ER-protein complexes and ER purification.
To prevent overgrowth of our culture by cells that had either lost the
ER gene through recombination, or inactivated the promoter driving ER
transcription, we used a bicistronic expression system (27), which
tightly couples expression of the ER and the antibiotic resistance
genes by transcribing them as a single bicistronic mRNA. Translation of
the bicistronic mRNA occurs from two different translation initiation
sites. HeLa-ER1, one of the stable HeLa-ER cell lines we developed
using the bicistronic mRNA system, expresses high levels of
biologically active epitope-tagged human ER.
Several proteins have been reported to enhance the binding of steroid receptors to their DNA response elements (28, 29). Recent studies have focused on high-mobility group protein 1, HMG-1. HMG-1 is a highly conserved nonhistone chromosomal protein that binds to DNA without exhibiting sequence specificity, but exhibits a strong binding preference for DNAs in nonlinear conformations (30). HMG-1 enhances the binding of purified progesterone receptor (PR) (31, 32), ER expressed in nonmammalian cells (11, 12), and ER DNA-binding domain (33) to their respective response elements. Recently, Edwards and co-workers (12) extended their earlier work on HMG-1 action and demonstrated that HMG-1 enhances transactivation in intact cells and DNA binding by purified estrogen, androgen, and glucocorticoid receptors. However, HMG-1 did not stimulate DNA binding by nonsteroid nuclear receptors. We used both crude nuclear extracts and purified FLAG epitope-tagged ER (fER) from the HeLa-ER1 cells to analyze the effect of HMG-1 on binding of the fER to the ERE. Since fER in crude HeLa-ER1 nuclear extracts and purified fER exhibited similar high-affinity binding to the ERE, it was surprising that addition of HMG-1 strongly stimulated ERE binding by both purified fER and by fER in crude HeLa-ER1 nuclear extracts. In transient transfections of HeLa cells and of MDA-MB-231 human breast cancer cells, HMG-1 elicited a dose-dependent stimulation of both 17ß-estradiol (E2)-dependent and E2-independent transactivation of an ERE-containing reporter gene. A high level of transfected HMG-1 expression plasmid increased transactivation by E2 by approximately 5-fold while transactivation by trans-hydroxytamoxifen was increased by 27-fold.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Recently, Kraus and Kadonaga (35)
also concluded that adding an N-terminal FLAG epitope does not alter
the properties of the ER.
|
), in which
expression of both the fER and the neomycin phosphotransferase gene are
under the control of a single promoter. The plasmid containing this
bicistronic construct was introduced into HeLa cells by
electroporation, and the cells were maintained under continuous G418
selection. Seven clones exhibiting reasonable growth rates were
isolated and analyzed for fER content. All seven of the G418-resistant
clones expressed significant levels of fER, with levels ranging from
20,000–1,300,000 molecules of fER per cell (Table 1). The cloned cell line with the highest
level of fER expression, designated HeLa-ER1, was chosen for more
detailed study. The HeLa-ER1 cell line shows no sign of the progressive
loss of fER expression we previously observed and has maintained its
high level of fER expression for more than 1 yr. This cell line
expresses about 20 times more ER than standard lines of MCF-7 human
breast cancer cells (Ref. 13 ; Table 1).
|
expressed in HeLa-ER1 cells
is functional in vivo, we determined whether it retained the
characteristics of biologically active ER. Functional liganded ER
should 1) show nuclear localization when ligand is present, 2) be toxic
to the cells at high expression levels, and 3) activate transcription
of an ERE-containing reporter gene. To ascertain the subcellular
location of fER in the presence and absence of E2, we
prepared nuclear and cytosol extracts and analyzed their ER content by
Western blotting. In the absence of E2, approximately half
of the fER was in the cytosol. Addition of E2 resulted in
complete nuclear localization of the fER (data not shown). Although the
level of fER in the HeLa-ER1 cells is quite high, most of the fER is in
a soluble and salt-extractable form, with only a small fraction in the
insoluble nuclear pellet.
To assess fER toxicity, cell growth was monitored in either the
presence or absence of 17ß-estradiol, or of the antiestrogens,
trans-hydroxytamoxifen (TOT) and ICI 182,780 (Fig. 2). 17ß-Estradiol had no effect on the
growth of wild-type HeLa cells, which lack ER (Table 1). TOT or ICI
182,870 also showed no effect on HeLa cell growth (data not shown).
Addition of 17ß-estradiol, or of the antiestrogen TOT, resulted in
rapid killing of the HeLa ER-1 cells. Estradiol and TOT were also toxic
to HeLa-ER2 cells (data not shown), which contain a much lower level of
ER than the HeLa-ER1 cells (see Table 1). In contrast, the pure
antiestrogen ICI 182,870 was not toxic to the HeLa-ER1 cells and
actually enhanced their growth (Fig. 2).
|
). The estradiol dose-response curve
exhibits saturation at 10-9 M, which is in
good agreement with the KD of the ER for E2
(see below). Even at the very high level of fER in the HeLa-ER1 cells,
there was no transcription of the reporter gene in the absence of added
E2. This finding is in agreement with our recent report
that under standard cell culture conditions transcription activation by
ER requires the presence of estrogen (36). TOT did not activate
transcription of the 2ERE-TK-CAT reporter gene and antagonized
E2-mediated transactivation.
|
, fER).
Western blotting with the ER-specific monoclonal antibody, H222 (37),
which recognizes an epitope near the C terminus of the ER and with the
anti-FLAG M2 antibody, which recognizes the N-terminal FLAG epitope,
revealed a single band comigrating with the mol wt standard at about 66
kDa (Fig. 4, B and C). The presence of a single band in the
silver-stained gel of pure fER and in Western blotting with antibodies
specific for N-terminal and C-terminal epitopes demonstrates that the
fER did not undergo significant proteolysis during purification.
|
). The crude and purified ER
exhibited KD values for 17ß-estradiol of 0.45–0.5
nM, which is in the range of affinities determined for
wild-type ER in various cell lines (38, 39). These data indicated that
other cellular proteins, which are removed on purification of the fER
to near homogeneity, do not influence its ability to bind ligand.
|
, H222, Anti-BSA), indicating it was an
ER-ERE complex. In addition, binding of ER to the ERE was competed by
increasing amounts of unlabeled ERE, but not by a 100- and 1000-fold
molar excess of unlabeled GRE/PRE (Fig. 6A, ERE and GRE). In protein
titrations, the purified fER exhibited a concentration-dependent
increase in binding to the ERE (Fig. 6A, fER). ERE binding by fER in
crude nuclear extracts and by purified fER was similar (Fig. 6B). Both
crude fER and purified fER exhibited high-affinity binding to the
ERE.
|
), with an affinity for the ERE
approximately 10-fold lower than that of purified fER from HeLa-ER1
cells (see below).
HMG-1 Enhances Binding of fER to the ERE
While our data showed that purified fER was capable of
high-affinity binding to the ERE, it did not address the question of
whether HMG-1 could enhance binding of fER to the ERE. To more directly
examine the role of HMG-1 in binding of fER to the ERE, we used the
method recently described by Jayaraman et al. (40) to
express and purify His6-tagged HMG-1 in E. coli.
The HMG-1 was more than 90% pure as judged by SDS-PAGE. To minimize
oxidation, the HMG-1 was stored in a buffer containing 1 mM
dithiothreitol (12, 40). Addition of HMG-1 resulted in a strong
stimulation of binding of purified fER to the ERE (Fig. 7A, fER). To quantitatively assess the
effect of HMG-1 on fER binding to the ERE, we titrated a constant
concentration of HMG-1 with increasing amounts of fER (Fig. 7B). In the
presence of purified HMG-1, purified fER exhibited a KD of
0.25 nM for binding to the ERE (Figs. 7B and 8), 3- to 4 times lower than the 0.8
nM KD we observed for purified fER in the
absence of HMG-1 (Fig. 8). Surprisingly, a clear stimulation of binding
was also observed when HMG-1 was added to crude HeLa-ER1 nuclear
extracts (Fig. 7A, NE). These data suggest that even in crude HeLa cell
nuclear extracts, in which HMG-1 has been shown to be present at high
levels (40), HMG-1 can be limiting for ER binding to the ERE. HMG-1
also enhanced binding of the commercially obtained purified
baculovirus-expressed ER to the ERE (data not shown), eliciting a
similar 3- to 4-fold increase in binding for both the purified fER and
the baculovirus-ER.
|
|
). These data provide the evidence that
HMG-1 levels are limiting for ER action in intact cells. These data
also show that transfected HMG-1 elicits a modest, but readily
detectable, level of ligand-independent transactivation by ER. Our
observation that HMG-1 elicited a low level of ligand-independent
transactivation by ER raised the possibility that HMG-1 might also
increase the agonist activity of an antiestrogen. To evaluate this
possibility, we examined the effect of HMG-1 on transactivation by TOT
in MDA-MB-231 cells, which have been used previously to study
antiestrogen action (42).
|
) and was almost 3 times greater than
transactivation by E2 in the absence of HMG-1 (Fig. 10). At
1 µg of transfected HMG-1, transactivation by TOT was approximately
60% of the level of transactivation observed with E2 at
the same level of HMG-1.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The 1,300,000 molecules of ER per cell expressed by the HeLa-ER1 cells is significantly lower than the level of ERval400 expressed using a regulated promoter system in CHO cells (13, 14). The CHO cell system expressing ERval400 will continue to find applications when the very highest levels of ER expression are required, while the HeLa-ER1 system should find applications in situations where an epitope-tagged ER is useful, or when a large-scale suspension culture of the ER-expressing cell line is required.
While our cell culture medium lacks known sources of estrogens, the
HeLa-ER1 cells still exhibit significantly slower growth than wild-type
HeLa cells (Fig. 2). That this results, at least in large part, from
expression of fER, and not from continuous selection for resistance to
G418, is shown by the increased growth rate of HeLa-ER1 cells when the
pure antiestrogen ICI 182,780 is added to the culture medium. However,
the mechanism by which the pure antiestrogen increases the growth rate
of the cells is unclear. ICI 182,780 and a closely related member of
this series of antiestrogens, ICI 164,384, have been reported to both
stimulate the degradation of ER and block its uptake into the cell
nucleus (46, 47). This raises the possibility that nuclear localization
of high levels of unliganded ER is somewhat toxic, even in the absence
of estrogen. Alternatively, the antiestrogen could be antagonizing the
effect of traces of estrogen remaining in our culture medium. However,
in the absence of added E2, there is no detectable
transcription of an ERE-containing reporter gene transfected into the
HeLa-ER1 cells (Fig. 3), indicating that there is not a significant
concentration of estrogen in the culture medium.
Interestingly, the antiestrogen TOT, which does not activate
transcription of a transfected ERE-containing reporter gene (Fig. 3),
kills the HeLa-ER1 cells even more efficiently than 17ß-estradiol.
Efficient killing of cells expressing ER by both 17ß-estradiol and
tamoxifen has been reported previously in CHO and mammary epithelial
cell lines (13, 14), suggesting that this is a general phenomenon.
Recent studies suggest that estrogens and TOT may kill breast cancer
cells by different mechanisms (48).
ER Expressed in HeLa-ER1 Cells Is Functional in Vivo
and in Vitro
While the high level of fER in the HeLa-ER1 cells (1,300,000
molecules ER/cell) contributes enormously to their utility as a source
of functional ER, the superphysiological level of ER in these cells
raises the related questions of whether the properties of this
recombinant fER are similar to those of native ER, and whether all or
most of the fER is actually functional. In vivo and in
vitro characterization of the fER demonstrates that its properties
are similar to those of naturally occurring ER. The liganded fER
displays nuclear localization as does ER in naturally occurring cells
(49). The fER efficiently activates transcription of an ERE-containing
reporter gene in response to nanomolar concentrations of
17ß-estradiol. Transactivation is completely dependent on the
presence of added 17ß-estradiol and is antagonized by the
antiestrogen TOT. This pattern of antiestrogen activity is similar to
that seen in both ER-positive cell lines and in cells transiently
transfected with an ER expression plasmid (36). In vitro,
the crude and purified fER exhibit an affinity for 17ß-estradiol
typical of ER from a variety of sources (38, 39).
It is difficult to unequivocally demonstrate that all of the fER is functional in vitro. However, comparison of the intensity of the bands seen in Western blots of fER, and of wild-type MCF-7 cell ER, with the amount of ligand bound by the two ER preparations suggests that most and perhaps all of the fER is competent to bind hormone. Comparison of the amount of nearly homogeneous fER with the amount of labeled 17ß-estradiol bound supports the view that most of the fER is competent to bind hormone. Analysis of the interaction of the purified fER with the ERE also supports the view that most of the fER is functional. Perhaps because different measurement techniques were used, there is considerable variation in reported KD values for binding of ER to the ERE (50, 51). The 0.8 nM KD we determined for purified fER is within the range of published values. If a large majority of the fER molecules were unable to bind DNA, we would be unlikely to observe such a low KD.
Recent reports indicated that purification of ER expressed in insect
cells using the baculovirus system resulted in loss of the ability to
bind to the ERE (11, 12). In agreement with other studies (10, 35), we
find that commercially obtained purified ER expressed in insect Sf9
cells binds with reasonably high affinity to the ERE. Using
electrophoretic mobility shift assays we determined a KD of
10 nM for ERE binding, which agrees with the 10
nM KD for the ERE determined by the commercial
supplier using a fluorescence assay. Our finding that ER expressed in
insect cells exhibits a 12-fold reduction in affinity relative to the
human ER expressed in human cells (Fig. 8) suggests that protein
folding or posttranslational processes, such as receptor
phosphorylation, may be different in the two systems. Since the
purified ER expressed in insect cells was obtained commercially and was
purified using a different method than we employed, it is also possible
that the difference in affinity seen with these receptor preparations
reflects differences in the preparation and purification of these
ERs.
HMG-1 Potentiates Binding of ER from HeLa-ER1 Cells to the ERE and
Is Limiting in HeLa Cells
To facilitate comparison to data obtained in crude nuclear
extracts, and to provide an environment that more nearly simulates
nuclear DNA, in which EREs are located in a large excess of nonspecific
DNA binding sites, our binding assays were done in 3 µg of
nonspecific DNA. Even in the presence of this large approximately
30,000-fold excess of nonspecific DNA, HMG-1 clearly stimulated ERE
binding by both purified fER and by fER in crude HeLa-ER1 cell nuclear
extracts. The KD of 0.25 nM for purified fER
binding to the ERE in the presence of HMG-1 is much lower than other
values obtained using gel mobility shift assays (50).
Our observation that adding HMG-1 to fER-containing nuclear extracts elicits a several fold increase in fER binding to the ERE explains our finding that purifying the fER does not reduce its affinity for the ERE. Because the conditions used for gel mobility shift assays differ from those in intact nuclei, we examined the ability of HMG-1 to stimulate transactivation by fER in vivo. In both HeLa cells and MDA-MB-231 cells, HMG-1 elicited a dose-dependent increase of several fold in transactivation by ER. Taken together, the ability of HMG-1 to stimulate binding of ER to the ERE in crude nuclear extracts and the ability of transfected HMG-1 to stimulate ER-dependent transactivation strongly suggest that, despite the presence of high levels of HMG-1 in HeLa cell nuclei (40), HMG-1 levels may be limiting for ER action. Our data support the proposal of Edwards and co-workers (12) that under at least some conditions HMG-1 levels might be limiting for steroid hormone action (12) and suggests that modulating either the intracellular level of HMG-1 or the number of free nuclear HMG-1 binding sites could alter ER binding to the ERE.
HMG-1 Strongly Enhances the Agonist Activity of TOT
In both HeLa cells and MDA-MB-231 cells, transfected HMG-1
elicited a dose-dependent increase of 3- to 5-fold in transactivation
by unliganded ER. These data are in agreement with our earlier
observation that an ER mutant exhibiting enhanced binding to the ERE
in vivo exhibited significant ability to activate
transcription in the absence of estrogen (52, 53). Because the fold
stimulation of transactivation by HMG-1 was similar for both unliganded
ER and for ER when E2 was added to the culture medium, it
remained possible that the activation of unliganded ER was actually due
to the presence of traces of estrogen in the culture medium. To
evaluate this possibility we examined the effect of HMG-1 on
transactivation by TOT. HMG-1 enhanced transactivation by TOT far more
effectively than it enhanced transactivation by E2. At 1
µg of transfected HMG-1 expression plasmid, transactivation by
E2 increased 5-fold while transactivation by TOT increased
27-fold, and transactivation by TOT was 62% of the level seen with
E2. Interestingly, an ER mutant exhibiting enhanced
affinity for the ERE also exhibited enhanced transactivation by TOT
(52, 53).
Studies from several laboratories have led to the concept that DNA binding can be one of several factors modulating the activity of steroid receptors (52, 53, 54, 55, 56). The ability of HMG-1 to enhance ligand-independent transactivation by ER and to increase the agonist activity of TOT supports earlier proposals that the ER gains the ability to activate transcription through an activation pathway. In this model, ligand binding, phosphorylation through signal transduction pathways, differential interaction with coactivators and corepressors, and DNA binding jointly contribute to the receptor’s transactivation potential. Thus, binding to the ERE without added HMG-1 can be thought of as moving the ER part way down its activation pathway, while enhanced binding of the ER to the ERE in the presence of added HMG-1 moves the receptor further down its activation pathway, decreasing the requirement for ligand and allowing TOT, normally a weak agonist, to show strong agonist activity. While our data are consistent with this provocative idea, the role of ERE binding in the changes that render the ER competent to activate transcription remains largely obscure.
Possible Mechanisms of HMG-1 Stimulation of Transactivation by
ER
Although HMG-1 has been reported to enhance sequence-specific
binding by P53 (40), PR (12, 31, 32), ER (11, 12), ER DNA binding
domain (33), and other steroid receptors including GR and AR (12), and
HOX family members (57), the mechanism by which HMG-1 enhances
sequence-specific DNA binding has not been clearly established. We find
that the electrophoretic mobility of fER-ERE complexes formed in the
presence or absence of HMG-1 is identical, suggesting a weak
interaction between HMG-1 and DNA, an interaction that is not
sufficiently stable to persist during electrophoresis. These data are
in agreement with earlier work in which HMG-1 did not alter the
electrophoretic mobility of protein-DNA complexes (11, 12, 31, 32, 33).
HMG-1 exhibits a strong preference for binding to nonlinear DNA. At limiting nuclear HMG-1 concentrations, HMG-1 would therefore be expected to preferentially bind to curved or bent DNAs. Some EREs are located in regions of the DNA likely to exhibit preferential binding by HMG-1. For example, the EREs in the Xenopus vitellogenin genes are located directly adjacent to a region of curved DNA (58, 59), and it is possible that binding of HMG-1 to this region of DNA facilitates binding of ER to the nearby vitellogenin EREs.
The most plausible explanation for the mechanism by which HMG-1 stimulates binding of ER to the ERE relates to its ability to induce or stabilize bending in DNA and its preferential binding to nonlinear or bent DNA. ER induces DNA bending upon binding to the ERE (60). Recently, we demonstrated that ER DNA binding domain preferentially binds to DNA bent in the same direction as the ER-induced DNA bend (61). It is possible that HMG-1 either distorts the DNA conformation to facilitate ER binding or, by its preferential binding to the bent DNA around the ER-ERE binding site, helps to stabilize the ER-induced bent DNA conformation.
Our data demonstrate that transfected HMG-1 can enhance transactivation by unliganded ER and by E2-ER and TOT-ER complexes in intact cells. While it is most likely that the ability of HMG-1 to facilitate binding to the ERE is responsible for these increases in transactivation by ER, other explanations have not been excluded. HMG-1 proteins appear to play a role in mediating assembly of nucleoprotein complexes (62), in chromatin decondensation (63), and in transcription by RNA polymerase II in vitro (64). Although effects of HMG-1 on chromatin assembly and disassembly seem less likely with the transiently transfected genes we used than for chromosomal genes, we cannot formally exclude the possibility that effects on chromatin structure and the basal transcription apparatus also contribute to the stimulation of ER-mediated transactivation by HMG-1.
In this work we describe the isolation of HeLa-ER cell lines using a generally applicable method for isolation of stable cell lines expressing a toxic protein. With the ability of HeLa cells to grow in large-scale suspension cultures, fER expression levels more than 10-fold higher than are seen in naturally occurring mammalian cells, and a simple one-step purification of the epitope-tagged fER, the HeLa-ER1 cells provide a useful complement to previously described ER-expressing cell lines (13, 14). The HeLa-ER1 system should find application in studies of ER-associated proteins and in biochemical and structural studies that require substantial quantities of highly purified, biologically active, mammalian ER. Our studies with fER from HeLa-ER1 cells reveal that nuclear HMG-1 levels appear to be limiting for ER action in vitro and demonstrate that in intact cells, HMG-1 strongly enhances the agonist potential of TOT.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Establishment of the Hela-ER Cell Lines
HeLa cells adapted for growth in DME/10%
charcoal-dex-tran-treated FBS were grown to 60–70% confluence in
100-mm culture plates. The cells were transfected with 20 µg of
SspI-linearized pIE DNA by electroporation (200 V, 1180
µFarad, low resistance using a Cell-Porter from GIBCO-BRL,
Gaithersburg, MD). The medium was replaced with selection media (50%
DME/10% charcoal-dextran-FBS, 50% conditioned medium + 1 mg/ml G418)
approximately 24 h post transfection. After 10 days, colonies were
isolated and reseeded into 24-well culture plates in selection media.
Cell lines were further expanded to confluence in standard growth
medium (without conditioned medium) in T175 flasks. For long-term
growth, the cells were maintained under selection in medium containing
0–200 µg/ml G418.
Transient Transfections and CAT Assays
Transient transfections in HeLa-ER1 cells were performed using
electroporation as described above for production of the stable cell
lines, with the exception that 20 µg of 2ERE-TK-CAT DNA and 4 µg of
cytomegalovirus (CMV)-luciferase as internal standard were used and the
media contained different concentrations of 17ß-estradiol
(10-11 M to 10-6 M)
and/or TOT (10-6 M, 10-7
M). After 48 h, the cells were harvested, and
mixed-phase CAT assays were performed as we have described (66).
Transient transfections of HeLa cells to compare transactivation by fER
and by wild-type ER were done by electroporation as described above
using the indicated quantities of CMV-ER or CMV-fER expression
plasmids, 20 µg of ATC4 (36), and 4.2 µg CMV-luciferase as internal
standard. The cells were maintained in 10-8 M
E2 for 48 h before harvesting. Transient transfections
in HeLa cells using HMG-1 were performed using Tfx-20 reagent (Promega,
Madison, WI), using the manufacturer’s protocol. Transfections were
done in six-well plates, containing 50 ng of CMV-ER, 50 ng of
CMV-luciferase as internal standard, 900 ng of 2ERE-TK-CAT, the
indicated amount of an HMG-1 expression plasmid (31, 41), and PTZ18 U
to bring the total amount of DNA to 1250 ng. The indicated
concentrations of hormone were added immediately after transfection.
After 48 h, the cells were harvested for CAT assays (66).
Transient transfection in MDA-MB-231 cells was performed using calcium
phosphate (42). Transfections were done in six-well plates containing
10 ng of CMV-ER, 50 ng of PRL-SV40 (from Promega) as internal standard,
1 µg of 4ERE-LUC (constructed by G. de Haan in this laboratory), the
indicated amounts of an HMG-1 expression plasmid, and PTZ 18U to bring
the total amount of DNA to 2.67 µg. Forty eight hours after the cells
had been shocked, the dual luciferase assay (Promega) was performed
using the manufacturer’s protocol.
Preparation of Cell Extracts and Immunopurification of
fER
Cells used for making nuclear and cytosol extracts were grown in
medium without added hormone. Extracts made in the presence of
estradiol were pretreated by resuspending the harvested cells (without
serum), adding E2 to 10-7 M, and
then incubating at 37 C for 30 min. Cell extracts were prepared as we
have described (67) except that after centrifugation to remove the
nuclear pellet we retained the cytosol. The nuclear resuspension buffer
was TEG 500 (50 mM Tris, pH 7.9, 0.1 mM EDTA,
0.5 mM EGTA, 0.5 M KCl, 1 mM
dithiothreitol, 50 µM ZnCl2, 10% glycerol,
50 ng/µl leupeptin, 5 ng/µl phenylmethylsulfonyl fluoride, 5
ng/µl pepstatin A, 0.5 ng/µl aprotinin), and the ammonium sulfate
precipitation and dialysis steps were omitted. For fER purification,
estradiol-treated HeLa-ER1 nuclear extracts were adjusted to 300
mM KCl and applied to an anti-FLAG epitope immunoaffinity
column (Anti-FLAG M2 affinity Gel, Eastman Kodak, Rochester, NY), at 50
µl of packed resin per ml of nuclear extract. The column was
subsequently washed 10 times with a total of 100 volumes of TEG300
containing 8 mM
(3-[(3-chloramidopropyl)dimethylammonio]-1-propane-sulfonate,
and the fER was eluted with FLAG peptide (N-DYKDDDDK-C, 0.2 mg/ml) in
TEG100.
Western Blots
Approximately 4 ng of purified fER and 20 µg of nuclear
extract containing or lacking fER were analyzed by electrophoresis on a
10% Glycine-SDS polyacrylamide gel, and the proteins were
electroblotted onto a nitrocellulose membrane. The membrane was probed
with the anti-FLAG M2 monoclonal antibody (at 1:2000 dilution) or
ER-specific primary antibody H222 at 0.12 µg/ml, incubated with
horseradish peroxidase-conjugated secondary antibodies (at 1:2000
dilution) and detected by chemiluminescence with the ECL kit (Amersham,
Arlington Heights, IL).
ER Ligand-Binding Assays
Whole-cell ER assays were carried out as described by Zhuang
et al. (66). In vitro estrogen-binding assays
were modified from the method of Carlson et al. (39). The ER
was diluted into binding buffer (50 mM Tris, pH 7.5, 10%
glycerol, 10 mM mercaptoethanol, 500 µg/ml BSA). The
bound ligand was assayed by adsorption onto hydroxyapatite for 15 min
at 4 C, followed by three washes with 1 ml 0.05 M Tris, pH
7.3. After the last wash, the pellet was resuspended in 0.5 ml of
ethanol and counted in 5 ml of scintillation fluid.
Electrophoretic Mobility Shift Assays
Electrophoretic mobility shift assays were carried out as
described previously (61) with some modifications. Briefly, end-labeled
ERE-containing probes (10,000 cpm/reaction) were combined with the
indicated amounts of purified ER, Hela-ER1 nuclear extract or HMG-1,
and 500 ng/µl BSA (or the amount of BSA required to reach 10 µg of
total protein when crude nuclear extracts were used); 3 µg of
poly-dIdC (Sigma) were present as nonspecific carrier DNA, 10%
glycerol, 75 mM KCl, 15 mM Tris-HCl, pH 7.9,
0.2 mM EDTA, and 0.4 mM dithiothreitol in a
volume of 20 µl and incubated at 25 C for 15 min. For antibody
supershift experiments and competition electrophoretic mobility shift
assays, conditions were the same, except for a preincubation for 10 min
on ice with either antibody (rabbit anti-BSA was a gift from S.
Miklasz, University of Illinois) or DNA competitor before addition of
the labeled ERE-containing probes. After probe addition, the reaction
mixtures were incubated at 25 C for 15 min and subjected to low ionic
strength 8% PAGE using a water jacket to maintain the gel at 4 C with
buffer recirculation. Gels were dried before autoradiography, and free
and bound forms of ERE and ER-ERE complex were quantitated by
PhosphorImager analysis (Molecular Dynamics, Sunnyvale, CA).
Determination of ER Concentrations and Calculation of
KD Values
The amount of ER in the ER preparations from HeLa cells and
baculovirus was determined by whole-cell ER assay as described above.
We also determined the actual amount of protein in purified fER
preparations (which migrate as a single band on SDS gels). To confirm
that the baculovirus and crude ER preparations contained the amount of
ER determined in the whole-cell ER assays, we used side-by-side Western
blots using purified fER as a standard and the ER-specific H222
monoclonal antibody to detect the ER.
We calculated KD values for binding of the ER preparations to the ERE essentially as described by Kim et al. (61). Briefly, the KD is the amount of ER required to up-shift 50% of the labeled ERE probe in a gel mobility shift assay. Under our gel shift conditions, essentially all of the labeled probe remained in the probe band, or was in a discrete up-shifted band.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
This research was supported by NIH Grants HD-16720 and DK-50080.
Received for publication July 27, 1998. Revision received December 29, 1998. Accepted for publication January 4, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. C. Roemer, J. Adelman, M. E. A. Churchill, and D. P. Edwards Mechanism of high-mobility group protein B enhancement of progesterone receptor sequence-specific DNA binding Nucleic Acids Res., June 1, 2008; 36(11): 3655 - 3666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Fink Neuropeptide Modulators of High Mobility Group Box 1 Secretion as Potential Therapeutic Agents for Severe Sepsis Am. J. Pathol., May 1, 2008; 172(5): 1171 - 1173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. D. Curtis, V. S. Likhite, I. X. McLeod, J. R. Yates, and A. M. Nardulli Interaction of the Tumor Metastasis Suppressor Nonmetastatic Protein 23 Homologue H1 and Estrogen Receptor {alpha} Alters Estrogen-Responsive Gene Expression Cancer Res., November 1, 2007; 67(21): 10600 - 10607. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Cheng, D. V. Yu, J.-H. Zhou, and D. J. Shapiro Tamoxifen Induction of CCAAT Enhancer-binding Protein {alpha} Is Required for Tamoxifen-induced Apoptosis J. Biol. Chem., October 19, 2007; 282(42): 30535 - 30543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Cheng, C. Zhang, and D. J. Shapiro A Functional Serine 118 Phosphorylation Site in Estrogen Receptor-{alpha} Is Required for Down-Regulation of Gene Expression by 17{beta}-Estradiol and 4-Hydroxytamoxifen Endocrinology, October 1, 2007; 148(10): 4634 - 4641. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Wang, C. Zhang, S. K. Nordeen, and D. J. Shapiro In Vitro Fluorescence Anisotropy Analysis of the Interaction of Full-length SRC1a with Estrogen Receptors {alpha} and beta Supports an Active Displacement Model for Coregulator Utilization J. Biol. Chem., February 2, 2007; 282(5): 2765 - 2775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Roemer, D. C. Donham, L. Sherman, V. H. Pon, D. P. Edwards, and M. E. A. Churchill Structure of the Progesterone Receptor-Deoxyribonucleic Acid Complex: Novel Interactions Required for Binding to Half-Site Response Elements Mol. Endocrinol., December 1, 2006; 20(12): 3042 - 3052. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. F. Eddy, S. Guo, E. G. Demicco, R. Romieu-Mourez, E. Landesman-Bollag, D. C. Seldin, and G. E. Sonenshein Inducible I{kappa}B Kinase/I{kappa}B Kinase {varepsilon} Expression Is Induced by CK2 and Promotes Aberrant Nuclear Factor-{kappa}B Activation in Breast Cancer Cells Cancer Res., December 15, 2005; 65(24): 11375 - 11383. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Das, R. C. Peterson, and W. M. Scovell High Mobility Group B Proteins Facilitate Strong Estrogen Receptor Binding to Classical and Half-Site Estrogen Response Elements and Relax Binding Selectivity Mol. Endocrinol., November 1, 2004; 18(11): 2616 - 2632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Loven, R. E. Davis, C. D. Curtis, N. Muster, J. R. Yates, and A. M. Nardulli A Novel Estrogen Receptor {alpha}-Associated Protein Alters Receptor-Deoxyribonucleic Acid Interactions and Represses Receptor-Mediated Transcription Mol. Endocrinol., November 1, 2004; 18(11): 2649 - 2659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. S. Melvin, C. Harrell, J. S. Adelman, W. L. Kraus, M. Churchill, and D. P. Edwards The Role of the C-terminal Extension (CTE) of the Estrogen Receptor {alpha} and {beta} DNA Binding Domain in DNA Binding and Interaction with HMGB J. Biol. Chem., April 9, 2004; 279(15): 14763 - 14771. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Krieg, S. A. Krieg, B. S. Ahn, and D. J. Shapiro Interplay between Estrogen Response Element Sequence and Ligands Controls in Vivo Binding of Estrogen Receptor to Regulated Genes J. Biol. Chem., February 6, 2004; 279(6): 5025 - 5034. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Amir, M. Barua, N. C. McKnight, S. Cheng, X. Yuan, and S. P. Balk A Direct {beta}-Catenin-independent Interaction between Androgen Receptor and T Cell Factor 4 J. Biol. Chem., August 15, 2003; 278(33): 30828 - 30834. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Loven, N. Muster, J. R. Yates, and A. M. Nardulli A Novel Estrogen Receptor {alpha}-Associated Protein, Template-Activating Factor I{beta}, Inhibits Acetylation and Transactivation Mol. Endocrinol., January 1, 2003; 17(1): 67 - 78. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Obrero, D. V. Yu, and D. J. Shapiro Estrogen Receptor-dependent and Estrogen Receptor-independent Pathways for Tamoxifen and 4-Hydroxytamoxifen-induced Programmed Cell Death J. Biol. Chem., November 15, 2002; 277(47): 45695 - 45703. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. S. Melvin, S. C. Roemer, M. E. A. Churchill, and D. P. Edwards The C-terminal Extension (CTE) of the Nuclear Hormone Receptor DNA Binding Domain Determines Interactions and Functional Response to the HMGB-1/-2 Co-regulatory Proteins J. Biol. Chem., July 5, 2002; 277(28): 25115 - 25124. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Yuan, M. L. Lu, T. Li, and S. P. Balk SRY Interacts with and Negatively Regulates Androgen Receptor Transcriptional Activity J. Biol. Chem., November 30, 2001; 276(49): 46647 - 46654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. WANG, H. YANG, C. J. CZURA, A. E. SAMA, and K. J. TRACEY HMGB1 as a Late Mediator of Lethal Systemic Inflammation Am. J. Respir. Crit. Care Med., November 15, 2001; 164(10): 1768 - 1773. [Full Text] [PDF]
|
|
|
|
|
|
S. A. Krieg, A. J. Krieg, and D. J. Shapiro A Unique Downstream Estrogen Responsive Unit Mediates Estrogen Induction of Proteinase Inhibitor-9, a Cellular Inhibitor of IL-1{beta}- Converting Enzyme (Caspase 1) Mol. Endocrinol., November 1, 2001; 15(11): 1971 - 1982. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Czura, Haichao Wang, and K. J. Tracey Dual roles for HMGB1: DNA binding and cytokine Innate Immunity, August 1, 2001; 7(4): 315 - 321. [Abstract] [PDF]
|
|
|
|
|
|
S. Oesterreich, P. Zhang, R. L. Guler, X. Sun, E. M. Curran, W. V. Welshons, C. K. Osborne, and A. V. Lee Re-expression of Estrogen Receptor {alpha} in Estrogen Receptor {alpha}-negative MCF-7 Cells Restores both Estrogen and Insulin-like Growth Factor-mediated Signaling and Growth Cancer Res., August 1, 2001; 61(15): 5771 - 5777. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Klinge Estrogen receptor interaction with estrogen response elements Nucleic Acids Res., July 15, 2001; 29(14): 2905 - 2919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Drew, D. C. Tang, P. E. Berg, and G. P. Rodgers The role of trans-acting factors and DNA-bending in the silencing of human {beta}-globin gene expression Nucleic Acids Res., July 15, 2000; 28(14): 2823 - 2830. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Mao and D. J. Shapiro A Histone Deacetylase Inhibitor Potentiates Estrogen Receptor Activation of a Stably Integrated Vitellogenin Promoter in HepG2 Cells Endocrinology, July 1, 2000; 141(7): 2361 - 2369. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. B. Ellwood, Y.-M. Yen, R. C. Johnson, and M. Carey Mechanism for Specificity by HMG-1 in Enhanceosome Assembly Mol. Cell. Biol., June 15, 2000; 20(12): 4359 - 4370. [Abstract] [Full Text]
|
|
|
|
|
|
D. Knutti, A. Kaul, and A. Kralli A Tissue-Specific Coactivator of Steroid Receptors, Identified in a Functional Genetic Screen Mol. Cell. Biol., April 1, 2000; 20(7): 2411 - 2422. [Abstract] [Full Text]
|
|
|
|
|
|
H. Kanamori, S. Krieg, C. Mao, V. A. Di Pippo, S. Wang, D. A. Zajchowski, and D. J. Shapiro Proteinase Inhibitor 9, an Inhibitor of Granzyme B-mediated Apoptosis, Is a Primary Estrogen-inducible Gene in Human Liver Cells J. Biol. Chem., February 25, 2000; 275(8): 5867 - 5873. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. C. Zhang and D. J. Shapiro Activation of the p38 Mitogen-activated Protein Kinase Pathway by Estrogen or by 4-Hydroxytamoxifen Is Coupled to Estrogen Receptor-induced Apoptosis J. Biol. Chem., January 7, 2000; 275(1): 479 - 486. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. He, C. H. Liang, and S. J. Lippard Steroid hormones induce HMG1 overexpression and sensitize breast cancer cells to cisplatin and carboplatin PNAS, May 23, 2000; 97(11): 5768 - 5772. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
